Add work_sets for better multi-gpu scaling

This replaces n slice buffers with a single master buffer and n
slices which reference into the master buffer. This allows a
single copy to upload or download the data.

@@ -978,6 +978,12 @@ class CyclesRenderSettings(bpy.types.PropertyGroup):
         min=8, max=8192,
     )
 
+    interleaved_slices: BoolProperty(

@@ -46,3 +46,3 @@
                                                       destination.num_components;
 
-  parallel_for(0, buffer_params.window_height, [&](int64_t y) {
+  /* Calculate how many full plus partial slices there are */

@@ -49,0 +65,4 @@
+
+  /* Copy over each slice to the destination */
+  parallel_for(0, slices, [&](int slice) {
+  //for(int slice = 0;slice < slices;++slice) {

@@ -254,0 +298,4 @@
+    slice_sizes[largest_weight] += leftover_scanlines;
+    slice_stride++;
+  } else if(leftover_scanlines < 0) {
+    VLOG_WARNING << "#######Used to many scanlines";

@@ -254,0 +301,4 @@
+    VLOG_WARNING << "#######Used to many scanlines";
+  }
+
+  VLOG_INFO << "===================SLICE allocatable:" << allocatable_slices << " fixed:"<< fixed_slices <<  "================";

@@ -412,2 +463,4 @@
   });
 
+  const double work_time = time_dt() - start_time;
+  VLOG(3) << "render time total for frame: "

@@ -78,2 +78,4 @@
       kg, state, render_buffer, scheduled_sample, tile->sample_offset);
 
+  /* Map the buffer coordinates to the image coordinates */
+  int tile_y = y - tile->slice_start_y;